JPS6243554B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to an apparatus for manufacturing solar cells. The present invention relates to a method and apparatus for manufacturing solar cells in which adjacent layers formed of amorphous semiconductor materials of different conductivity types are deposited in separate and separate glow discharge deposition chambers.

A photovoltaic device is known as a structure that converts sunlight into usable electrical energy. One example of this type of device is an amorphous silicon solar cell made of a multilayer structure of amorphous silicon to which impurities are added. A method for continuously depositing doped layers for such amorphous silicon solar cells and cell structures in a glow discharge chamber is described in US Pat. No. 4,226,898.

However, as described in this patent, the doped layer and the intrinsic layer of amorphous material are formed in a single vacuum chamber within a manufacturing device. In the case of this production equipment, a reaction gas mixture containing various doping materials (in case of forming an impurity doped layer)
and the reactant gas mixture without doping material (if forming an intrinsic layer) via a number of conduits.
are introduced sequentially into a single chamber.

A single chamber batch processing approach may limit the optimization and production rate of the final cell structure more than desired. The fabrication of solar cells consisting of multilayer structures with adjacent layers (including intrinsic layers) made of different types of materials in a single glow discharge chamber (deposition chamber) requires complex control equipment and time constraints. Such technology is required. In particular, the repeated depressurization and heating and cooling after each layer is deposited to produce each battery greatly increases the average time to produce the battery.

Furthermore, contamination of layers made of different types of materials, especially intrinsic layers, due to undesired processing or other factors must be avoided for optimal operation of the cell. In order to do so, in the case of a single chamber system, it is necessary to perform evacuation midway to avoid cross-contamination.

The above-mentioned and other disadvantages of the prior art are overcome by the present invention, in which adjacent layers of amorphous silicon having different electrical properties are deposited on a substrate in separate environmentally isolated glow discharge regions. It can be overcome by depositing.
These separate regions each contain a predetermined reaction gas mixture and may be a plurality of adjacent deposition chambers separated from each other to avoid cross-contamination. The substrates are sequentially advanced or transported through separate areas or chambers and adjacent layers having different electrical properties required for the particular cell configuration are deposited. The substrate may be a continuous web, such as stainless steel, which is fed substantially continuously into separate areas or chambers, and each layer is deposited to obtain the desired cell structure. A mask may be used if required for the particular cell configuration.

Next, a preferred specific example of the present invention will be explained based on the drawings.

FIG. 1 shows various steps in one embodiment of a continuous solar cell production system according to the present invention. Substrate 10 is comprised of any desired material on which amorphous silicon may be deposited, and may be transparent or opaque to incident sunlight. The substrate 10 may also be an individual plate, such as a metal foil, metal, glass or polymer, carried by a web or web-like transport mechanism. If it is a metal or polymer, such as stainless steel or aluminum, the web can be fed from a semi-continuous source, such as a large roll.

When the substrate 10 is made of a continuous web, sprocket holes are bored in the substrate 10 by a punching machine 12 that forms sprocket holes along both side edges of the web-like substrate 10, and the sprocket holes are formed in the substrate 1 through the sprocket holes.
0 longitudinally and provides a longitudinal reference mark that longitudinally integrates subsequent steps.
Of course, the punch 12 and sprocket holes may not be used, and edge guides or other alignment devices may be used.

After drilling, the substrate 10, if it is made of aluminum, is transported to an anodizing bath where a fourth insulating layer 16 of aluminum oxide is deposited on the substrate 10, in particular on the surface on which the deposition will take place. (see figure) is formed. If stainless steel is used as the substrate 10 and an insulating layer is desired, for example
SiO ₂ , Si ₃ N ₄ or the like may be deposited on the substrate 10 .

A series of base contacts are then optionally formed on the insulating layer. The base contact can be longitudinally aligned with the sprocket holes so that the base contact can be properly positioned to perform a sequence of operations of the device. Two of these base contacts are shown at 18 and 20 in FIG. Depending on the series or parallel connection arrangement required for a given cell, the orientation of the base contacts can be chosen as shown in FIG. 1 or differently from FIG. The apparatus 22 for forming this base contact is conventional and typically involves application of a mechanical or lithographic mask, subsequent formation of the base contact, and subsequent mask removal. The actual formation of the base contact can be carried out by methods known in the art, such as vapor deposition, sputtering, silk screening, printing, etc., and a detailed explanation thereof will not be necessary for those skilled in the art.

The conductive substrate 10 can be used as a common electrode as it is without forming an insulating layer and a base contact, and therefore the step of forming the insulating layer and the base contact and the masking step can be omitted. In this case, all the cells are connected in parallel with a substrate 10 serving as a common electrode (see FIG. 5). Conversely, when a glass substrate or a polymer substrate is used as the substrate 10, it is not necessary to form an insulating layer.

Drilling, anodizing, and base contact formation, if performed, may be performed sequentially by equipment arranged sequentially to act on the same moving substrate, but each step Each step may be performed separately using separate equipment, with the continuous web-like substrate being wound up each time the step is completed.

The deposition or adhesion of amorphous silicon onto the substrate 10 is carried out in the deposition chambers 24 and 2 shown in FIGS.
It will be held on 6.28. An example of the interior of the deposition chamber 24 is shown in FIG. Although three separate deposition chambers are shown in FIG. 2, one large chamber can be suitably partitioned into individual deposition or deposition regions, each of which is of a particular conductivity type (e.g., n-type). , p-type, or intrinsic) amorphous silicon.

Each deposition area is defined by the length of the chamber or the number of chamber modules depending on the thickness of the deposited layer and the rate of deposition. All deposition areas are separated from each other. This deposition device 2
4, 26, and 28 respectively form p-type, intrinsic and n-type layers of amorphous silicon (or n-type, intrinsic and p-type layers) from plasmas of individual reactant gas mixtures.
(form layer) is deposited by glow discharge. By depositing each layer separately, a battery consisting of an amorphous silicon layer with good electrical properties is obtained.

After depositing the amorphous silicon layer, a top contact layer 30 is placed on top of the amorphous silicon layer for collecting the current generated in the photovoltaic cell.
(FIG. 4). This layer 30
is made of a transparent material in order to pass the applied light energy to the silicon layer of the solar cell when the substrate 10 is opaque.

Commonly used transparent conductive materials are indium tin oxide, tin oxide or indium oxide.
For cells formed on transparent substrates, the structure may include a transparent conductive oxide (TCO) on the substrate and an opaque contact on the top layer. In most cases, the TCO layer is not conductive enough to collect current from a large area of the cell, so a current collection grid made of a suitable metal is used along with the TCO, as is well known to those skilled in the art. used. If each cell is electrically isolated (not connected in parallel by a common layer), a metal connection layer 31 can be further deposited to connect the individual cells in series or parallel (Fig. 4). ).

Since the amorphous silicon layer is highly reflective of visible sunlight, much of the incident energy would normally be reflected. To prevent this energy loss, an antireflection (AR) layer 32 is formed (FIG. 6). This AR layer reduces the amount of light reflected. The AR layer can be formed from insulating materials such as zinc sulfide, zirconium oxide, silicon nitride, and titanium oxide. However, if TCO is used as a top contact layer, the thickness of the TCO layer can be chosen such that the TCO layer functions as a top contact and as an AR layer. This simplifies the battery structure and manufacturing process. Deposition device 3 shown in FIG.
4 is a top contact layer 30 and an AR layer 32
are deposited if these layers 30, 32 are used.

Although the deposition of layers 30 and 32 completes the formation of the solar cell structure, additional lamination is desirable to protect the cell structure from physical damage. A laminator 36 applies protective webs 38, 40 to the front and back sides of the substrate on which all the elements of the solar cell structure are formed.

Once this lamination process is completed, the solar cell can be connected to the outside, and the web-like substrate is
It is disconnected as required to supply the desired voltage and current. In this way, continuous strips can be formed and economical production of solar cells can be carried out.

An important aspect of the invention is that the deposition of separate amorphous silicon layers takes place in separate deposition chambers 24, 26, and 28, as schematically shown in FIG. Three separate deposition chambers are shown for sequentially depositing a p-type amorphous silicon layer 42, an intrinsic amorphous silicon layer 44, and an n-type amorphous silicon layer 46 (FIG. 4). As mentioned above, chambers 24, 26,
28 are separated from each other to avoid mixing of the components of the individual reaction gas mixtures. The amorphous silicon layer is deposited in the reverse order,
The n-type layer 46, the intrinsic layer 44, and the p-type layer 42 may be formed in this order.

The layer arrangement shown in FIG. 4 is for light incident from the top. When a transparent substrate is used instead of the opaque substrate 10, light can be incident from the substrate 10 side (FIG. 6). Additionally, a shot barrier or M-IS can be used if desired (FIG. 7).

In the above, the number and length of the deposition chambers or regions, their location along the production line and the material to be deposited can be selected for each desired solar cell structure.

FIG. 3 shows an example of the deposition chamber 24 in more detail. In FIG. 3, the substrate 10 is moved toward the viewer (towards the viewer). Housing 48 surrounds deposition chamber 24 and has an inlet/outlet for substantially continuous entry and exit of substrate 10, as described below. The heater 50 is the substrate 1
It may be a large area infrared heater located near zero (FIG. 3). Deposition occurs on the opposite surface of substrate 10. The heating of the substrate and its temperature control are described in the patent application filed in the United States by Robert F. Edgerton at the same time as this case. It can be carried out by the method and apparatus described in the US application entitled ``Tensual Plasma Deposition Process''.

For example, the processing preparation gas may be applied to the substrate 10.
The substrate is deposited from a pair of manifolds 52, 52 having a plurality of openings that direct the flow along one surface (the surface to be deposited) toward the center of the substrate and perpendicular to the direction of movement of the substrate. 10 is supplied to the deposition surface side. Alternatively, the reactant gas may be supplied to the United States at the same time as the present case, such as Masatsugu Iss, Timothy J. Bernard (Timothy J.
Barnard and David A. Guttuso, entitled ``Cathode for Generating a Plasma''. can be introduced.

The gases supplied to the reaction chamber are preferably SiF ₄ and hydrogen, and argon or other gases such as U.S. Pat.
An inert diluent gas, such as the inert gas described in the U.S. application entitled "Improved Method for Plasma Deposition of Amorphous Materials," filed by Cannella and Masatsugu Is. May contain.

Uniform gas flow is desired and therefore a large number of openings are preferably formed in the manifolds 52,52. As a result, the gas can be directed substantially parallel to and adjacent to the deposition side surface of the substrate 10.

Exhaust port 56 is connected to a vacuum pump (not shown) to evacuate spent gas and maintain pressure balance. An electrode 58 is installed at a distance from the substrate 10, and the electrode 58
Plasma is generated between the substrate 10 and the substrate 10 . Gas is exhausted through electrode 58, preferably through a plurality of apertures 60 formed in electrode 58, and a uniform flow is maintained. In the plasma, the processing gas is mainly a silicon fluoride-hydrogen gas mixture, including various species such as SiF ₄ , SiF ₃ , SiF ₂ , SiF and other species containing hydrogen such as SiHF, SiHF ₂ , SiHF ₃ etc. Furthermore, if necessary, a dope component well known in the art is included. According to those skilled in the art, some of these species are transitional. The degree of vacuum or reduced pressure achieved at exhaust port 56 is such as to provide a pressure such that a glow discharge plasma can be maintained at the surface of substrate 10. Pressures in the range of 0.1 to 3 Torr are preferred.

Although the substrate 10 is grounded, the electrode 58 is connected to a power source 62 that provides electrical energy to generate and maintain a glow discharge plasma in the vicinity of the substrate 10 that deposits the amorphous silicon layer. Power supply 62 is typically an AC power supply operating in the radio frequency range, but may also be a DC power supply operating at a voltage that produces a glow discharge plasma. If radio frequency power is desired, the power supply may operate at low power from 50 to 200 kilohertz, for example as described in the third-noted US application. In addition to providing power to generate the glow discharge plasma, power supply 62 can apply a DC bias between electrode 58 and substrate 10 to control substrate bias. The DC bias applied to the plasma provides better control over the deposition process of amorphous silicon from the plasma.

When fabricating discrete or strip cells, it may be necessary to mask the surface of the substrate so that the plasma deposits amorphous silicon only on the desired areas. This mask may be realized by an endless mask belt 64 (FIG. 2) that moves with the substrate 10 in close proximity to the surface of the substrate 10. Alignment can be accomplished by holes formed in the side edges of the substrate so that the mask 64 can be properly positioned relative to the substrate 10. The strip cells can be oriented parallel to the direction of travel of the substrate through each deposition chamber, in which case no longitudinal alignment is required. Mask 64 is a continuous strip mask, and guide roll 65 within housing 48
(Figure 2). A portion 63 of the mask belt 64 that is located on the lower side and does not function directly
(FIG. 2) may be located below the electrode 58. Since the mask belt 64 has a large open area,
The flow of exhaust gas from the exhaust port 56 to the vacuum pump is not obstructed.

Each of the deposition chambers 24, 26, 28 is formed in the same way, and each of the deposition chambers 26, 28 also has a substrate 1.
Mask belts 66, 68 moving in the forward direction of 0
have. Although each deposition chamber 24, 26, 28 may be of the same construction, the composition of the plasma generated in each chamber is somewhat different due to the differences in the type of layer deposited within each chamber. The gas supplied to the manifold may be different for each deposition chamber, but the gas supplied to each deposition chamber is the same and the doping gas, e.g. phosphine (PH ₃ ) to provide n-type or p-type, is added elsewhere. Alternatively, diborane (B ₂ H ₆ ) may be supplied. for example,
A separate source of dope gas in an inert gas such as argon can be provided. It is desirable that the flow of the gas supplied to the deposition surface of the substrate 10 be uniform, so if a separate source of a mixed gas of dope gas and inert gas is provided, the gas will flow through the manifolds 52, 5.
It is preferable to mix the liquids before they are supplied into the manifolds 2 and discharged from the openings of the manifolds 52, 52.

Substrate 10 in each deposition chamber 24, 26, 28
The residence time of depends on the deposition rate and thickness of the layer to be deposited. For example, when making a PIN device, the thickness of each layer is 50 to 200 Å,
It can be 2000-6000 Å and 100-500 Å. That is, if the substrate consists of a continuous web and the deposition rate in each chamber is the same, the different deposition areas will have lengths proportional to the thickness to be deposited. The thickness of the deposited layer is described, for example, in a U.S. application filed at the same time as this case by Robert F. Edgerton under the title ``Optical Methods for Controlling Layer Thickness.'' can be monitored and controlled by the methods and equipment provided.

Deposition chambers 24, 26, 28 for supply gas and dope gas for plasma deposition of amorphous silicon
Control devices 70, 72, and 74 are connected to each of them to control process variables within. The vacuum pump is also controlled to maintain proper pressure levels to maintain proper plasma discharge balance, and the heater temperature is also controlled. In this way, continuous production can be performed. This system can be operated by slowly advancing the substrate in a continuous manner or by cycling the required portions of the substrate from one step to the next.

Each deposition chamber must have a controlled gas atmosphere that can provide precise conditions for depositing amorphous silicon and achieve appropriate impurity doping levels. The slits through which the substrate 10 enters and exits each deposition chamber are narrow;
However, separation means are also required to prevent intermixing and contamination of gases in adjacent deposition chambers. This separating means has a narrow slit through which the substrate 10 can pass, and the separating device 7 surrounds the substrate.
It consists of 6,78. The slit or central hole in each separator may be evacuated or flushed with argon or other inert gas, in which case the substrate 10 passes through the separator. All reactant gases are removed from the part.

The above-mentioned separation device is, for example, the separation valve described in the U.S. patent filed under the name "Isolation Valve" by Masatsugu I. and David A. Gattuso in the United States at the same time as this case. It may be hot. In FIG. 2, it is shown that the web-shaped substrate 10 is supplied and withdrawn within a vacuum chamber, but in the case of a completely continuous system, the substrate would enter from another process.
The deposition chamber 24 will exit to another process.
Separation devices would also be required at the inlet of the chamber 28 and the outlet of the chamber 28.

In the apparatus configured as described above, the deposition chamber 24,
Separation of the deposition atmosphere between 26 and 28 was achieved, with controlled equilibrium within each deposition chamber by continuous inflow of amorphous silicon deposition feed gas and controlled inflow of dope gas, as well as vacuum evacuation of spent reactant gases. A stable operation is maintained and stable plasma and deposition conditions are maintained within each deposition chamber.

FIGS. 4-7 show four examples of solar cells made according to the present invention. Fourth
The figure shows a solar cell consisting of a plurality of PIN cells 80. The substrate 10 may be made of metal or an insulator. The cells 80 may be formed in strips separated from each other by the masks described above. The insulating layer 16 is the metal substrate 1
0, but this may be omitted if the substrate is an insulator.

Multiple base contacts (2 of which are 1
8 and 20) are deposited on the insulating layer 16. The subsequent deposition is the same on each contact, starting with the p-type layer 42 and then the intrinsic layer 44.
and n-type layer 46 is deposited. A top contact layer 30 such as indium-tin oxide is deposited and optionally an AR layer.
The cell may include a grid 82 for collecting current, which grid 82 may be electrically connected to other cells, for example by a connecting metal layer 31, if desired. The entire cell is then protected and surrounded by laminate layers 38 and 40.

FIG. 5 shows a PIN type solar cell device 84 according to the second embodiment. In this device, a p-type layer 42, an intrinsic layer 44, and an n-type layer 46 are deposited over a metal substrate 10.
Individual batteries 86 are connected in parallel and masked or
Defined by photolithography of TCO layer 30. The battery 86 is connected to the grid 8 for current collection.
2', and this grid 82' can be connected as desired.

FIG. 6 shows a PIN type battery device 88 according to a third embodiment having a transparent substrate such as glass. In this case, sunlight is shown to be incident from the substrate 10 side. On the board 10
AR layer 32 is formed, followed by grid 90, if desired. Next, p-type layer 4
2. An intrinsic layer 44 and an n-type layer 46 are formed. Since the p-type layer 42 is used on the light incident side,
It is desirable to provide an intermediate layer between TCO 30 and p-type layer 42 to improve electrical compatibility therebetween. Finally, parallel bottom conductors 92 are deposited on top layer 46 in the desired pattern.

In FIG. 7, an M-I-S type device 94 is shown. An n-type layer 46 and an intrinsic layer 44 are formed on the metal substrate 10. An insulating layer 96 is formed over layer 44, and active metal contacts 98 are then formed on the individual cells. An AR layer 32 can be formed on the contact 98.

[Brief explanation of the drawing]

FIG. 1 is a diagram showing an example of the solar cell manufacturing process of the present invention, FIG. 2 is an explanatory diagram of a solar cell manufacturing apparatus according to a preferred embodiment of the invention, and FIG. 3 is a diagram showing the configuration of the deposition chamber of FIG. 2. The partially cutaway figures and FIGS. 4 to 7 are explanatory cross-sectional views of the solar cell structure obtained according to the present invention. 10... Substrate, 24, 26, 28... Glow discharge deposition chamber device, 42... P-type amorphous silicon layer, 44... Intrinsic amorphous silicon layer, 4
6...n-type amorphous silicon layer, 56...exhaust port, 58...electrode, 62...power supply, 64,
66, 68...mask, 52...manifold,
76, 78...separation device.

Claims (1)

[Scope of Claims] 1. A substrate advancing device that continuously moves a strip-shaped substrate so as to continuously advance the substrate in the longitudinal direction of the substrate; and a substrate advancing device that moves continuously in the longitudinal direction under the control of the advancing device. A first amorphous semiconductor layer of one of n-type and p-type conductivity is continuously deposited by glow discharge on one surface of the strip-shaped substrate. a first glow discharge deposition device having a first glow discharge deposition chamber; and a slit through which the strip substrate is continuously passed under the control of the advance device, the substrate being communicated with the first glow discharge deposition chamber. , n-type and p-type semiconductor layers are deposited on the first amorphous semiconductor layer on the strip substrate that is continuously moving in the longitudinal direction under the control of the advancing device.
a second glow discharge deposition apparatus having a second glow discharge deposition chamber configured to continuously glow discharge deposit a second amorphous semiconductor layer of the other conductivity type; While glow discharge deposition is being carried out continuously in each of the glow discharge deposition chambers,
When the strip substrate is continuously moving through the slit between the first glow discharge deposition chamber and the second glow discharge deposition chamber, the first deposition chamber and the second deposition chamber are connected to the peripheral wall of the slit. and separation means for preventing the deposition atmosphere in the first deposition chamber from being contaminated by the deposition atmosphere in the second deposition chamber in the slit communicating with the substrate through a gap between the amorphous solar cell and the substrate. Manufacturing equipment. 2. The manufacturing apparatus according to claim 1, wherein each of the first and second deposition apparatuses has a mask that allows each of the first and second layers to be deposited in a predetermined shape. . 3. The manufacturing apparatus according to claim 2, wherein the mask comprises an endless belt having a surface that moves in the same direction as the strip substrate in the vicinity of the strip substrate. 4. Each of the first and second deposition devices includes a manifold on each side of the substrate defining a plurality of openings configured to direct a reactant gas in close proximity to the one surface of the substrate. The manufacturing apparatus according to any one of claims 1 to 3, which are adjacent to each other. 5. The opening of the manifold is configured to direct the reactant gas from the side edges of the substrate toward the center and in a direction substantially perpendicular to the direction of movement of the substrate, and the opening of the manifold each from the manifold so that the flow of reactant gas is substantially symmetrical with respect to the substrate to achieve equilibrium between the flow of new reactant gas to the substrate and the removal of spent gas from the substrate. 5. The manufacturing apparatus of claim 4, having a symmetrical set of exhaust ports in the deposition chamber. 6. Claims 1 to 5, wherein each of the first and second deposition devices has an electrode adjacent to the one surface of the substrate and connected to a power source to generate plasma. Manufacturing equipment according to any of the above. 7. The manufacturing apparatus according to claim 6, wherein the power source is configured to apply a DC control bias to the electrode to the substrate. 8. A substrate advancing device that continuously moves the substrate so as to continuously advance the strip substrate in the longitudinal direction of the substrate; and one of the strip substrates that moves continuously in the longitudinal direction under the control of the advancing device. A first glow discharge deposition layer provided along the traveling path of the strip-shaped substrate in order to continuously glow discharge deposit a first amorphous semiconductor layer of one of the n-type and p-type conductivity on the surface of the substrate. a first glow discharge deposition device having a chamber; and a first glow discharge deposition chamber connected to the first glow discharge deposition chamber through a first slit through which the strip substrate is continuously passed under the control of the advancement device; a second layer configured to continuously glow discharge deposit a substantially intrinsic amorphous semiconductor layer onto the first amorphous semiconductor layer on the strip substrate moving continuously in the longitudinal direction under the control of the apparatus; a second glow discharge deposition device having a glow discharge deposition chamber of; and a second glow discharge deposition device communicating with the second glow discharge deposition chamber via a second slit through which the strip substrate is continuously passed under the control of the advancing device. and depositing a third amorphous semiconductor layer of the other conductivity type of n-type and p-type on the second amorphous semiconductor layer on the strip substrate that is continuously moving in the longitudinal direction under the control of the advancing device. a third glow discharge deposition apparatus having a third glow discharge deposition chamber configured to perform continuous glow discharge deposition; and a third glow discharge deposition device configured to perform glow discharge deposition continuously in each of the first and second glow discharge deposition chambers. While it is being done
When the strip substrate is continuously moving through the first slit between the first glow discharge deposition chamber and the second glow discharge deposition chamber, the first deposition chamber and the second deposition chamber are connected to each other. The deposition atmosphere in the first deposition chamber is prevented from being contaminated by the deposition atmosphere in the second deposition chamber in the first slit, which is communicated with the peripheral wall of the first slit through the gap between the substrate and the first slit. while glow discharge deposition is being continuously performed in each of the second and third glow discharge deposition chambers;
When the strip substrate is continuously moved through the second slit between the second glow discharge deposition chamber and the third glow discharge deposition chamber, the second deposition chamber and the third deposition chamber are connected to each other. The deposition atmosphere in the second deposition chamber is prevented from being contaminated by the deposition atmosphere in the third deposition chamber in the second slit, which is communicated with the peripheral wall of the second slit through the gap between the substrate and the second slit. an amorphous solar cell manufacturing apparatus having a second separation means for preventing 9. Claims in which each of the first, second, and third deposition devices has a mask that allows each of the first, second, and third layers to be deposited in a predetermined shape. The manufacturing apparatus according to item 8. 10. The manufacturing apparatus according to claim 9, wherein the mask comprises an endless belt having a surface that moves in the same direction as the strip substrate in the vicinity of the strip substrate. 11 Each of the first, second, and third deposition devices includes a respective manifold defining a plurality of openings configured to direct a reactant gas in close proximity to the one surface of the substrate. The manufacturing apparatus according to any one of claims 8 to 10, wherein the manufacturing apparatus is provided adjacent to each side edge of the substrate. 12 The opening of the manifold is configured to guide the reactive gas from the side edge of the substrate toward the center and in a direction substantially perpendicular to the direction of movement of the substrate, and the opening of the manifold Each of the deposition devices has substantially the same flow from the manifold so that the flow of reactant gas is substantially symmetrical with respect to the substrate to achieve equilibrium between the flow of fresh reactant gas to the substrate and the removal of spent gas from the substrate. 12. The manufacturing apparatus according to claim 11, further comprising a set of symmetrically symmetrical exhaust ports in the deposition chamber. 13. The manufacturing apparatus according to any one of claims 8 to 12, wherein the reaction gas introduced into each of the first and third deposition apparatuses includes a silicon-containing gas and a dope gas. 14. The manufacturing apparatus according to claim 13, wherein the reaction gas is mixed with argon. 15. The manufacturing apparatus according to claim 13 _, wherein the dope gas introduced into one of the first and third deposition apparatuses is _B2H6 . 16. The manufacturing apparatus according to claim 15, wherein the dope gas introduced into the other of the first and third deposition apparatuses is _PH3 . 17. Claim 8, wherein each of the first, second and third deposition devices has an electrode adjacent to the one surface of the substrate connected to a power source to generate plasma. 17. The manufacturing apparatus according to any one of Item 16. 18. The apparatus of claim 17, wherein the power source is configured to provide a substrate with a DC controlled bias to the electrode.